Volume 11, Issue 2, Pages 113-124 (February 2010) Fyn-Dependent Regulation of Energy Expenditure and Body Weight Is Mediated by Tyrosine Phosphorylation of LKB1  Eijiro Yamada, Jeffrey E. Pessin, Irwin J. Kurland, Gary J. Schwartz, Claire C. Bastie  Cell Metabolism  Volume 11, Issue 2, Pages 113-124 (February 2010) DOI: 10.1016/j.cmet.2009.12.010 Copyright © 2010 Elsevier Inc. Terms and Conditions

Figure 1 Acute Pharmacological Inhibition of Fyn Increases Energy Expenditure C57BL6/J males received an injection of vehicle or SU6656 (4 mg/kg) at 0700 hr and were placed into metabolic chambers without access to food. Respiratory quotient (RQ) and oxygen consumption (VO2) were recorded during the dark period preceding the injection and during the light period following the injection. (A) Shown is respiratory quotient (RQ) from the average of four mice injected with vehicle (open circles) and four mice injected with SU6656 (dark circles). (B) Shown is VO2 recorded before (dark period) and after the injection (light period). (C) Energy expenditure (EE) was calculated using the equation of Weir, EE (Kcal/kg/hr) = (3.815 × VO2) + (1.232 × VO2). (D) Physical activity was recorded during the dark (before injection) and light (after injection) periods; 0.001 < p < 0.01. Cell Metabolism 2010 11, 113-124DOI: (10.1016/j.cmet.2009.12.010) Copyright © 2010 Elsevier Inc. Terms and Conditions

Figure 2 SU6656-Induced Fyn Inhibition Promotes Fat Mass Loss (A) Body weight distribution of vehicle and SU6656-injected mice before (T = 0) and after (T = 12 hr) the injection. (B) Total weight loss 12 hr after the injection of vehicle (open bar) and SU6656 (dark bar) treated animals. (C) Fat mass before (T = 0) and after (T = 12 hr) vehicle or SU6656 injection. (D) Lean mass before (T = 0) and after (T = 12 hr) vehicle or SU6656 injection (see also Figure S1). Results are expressed as mean ± SEM. Cell Metabolism 2010 11, 113-124DOI: (10.1016/j.cmet.2009.12.010) Copyright © 2010 Elsevier Inc. Terms and Conditions

Figure 3 Fyn-Specific Inhibition Increased Skeletal Muscle Fatty Acid Oxidation and T172 AMPK Phosphorylation (A and B) Palmitate oxidation was determined in (A) red muscle (soleus) and (B) white gastrocnemius muscle (White Gastroc.) of mice injected with vehicle or SU6656 (4 mg/kg). Data are the mean ± SE of five independent experiments. (C) Shown are phospho-T(172)-AMPK and total AMPK protein expression levels in white gastrocnemius of vehicle or SU6656-treated mice. (D) Shown are phospho-ACC and total ACC protein expression levels in white gastrocnemius of vehicle or SU6656-treated mice. (E) Three-month-old Fyn knockout (diamonds) mice and their controls (circles) were treated with vehicle (open symbols) or SU6656 (filled symbols) at the beginning of the light cycle. Respiratory quotient (RQ) was recorded during the preceding dark period and during the 12 hr following the injection. (F) Shown are expression levels of Src and Lyn kinase in white adipose tissue (WAT), liver, gastrocnemius (Gastroc), and soleus muscle of Fyn null mice (Fyn−KO) and their controls (WT). Cell Metabolism 2010 11, 113-124DOI: (10.1016/j.cmet.2009.12.010) Copyright © 2010 Elsevier Inc. Terms and Conditions

Figure 4 Fyn Kinase Activity Regulates LKB1 Subcellular Distribution (A) C2C12 myotubes were transfected with pEGFP-LKB1 and incubated with vehicle or SU6656 (10 μM) for 2 hr. Cells were fixed and mounted with proQ diamond with DAPI solution. (B) Shown is percentage of cells with pEGFP-LKB1 signal detected in the cytoplasm. (C) C2C12 myotubes were cotransfected with pEGFP-LKB1 and pcDNA3-Fyn-KD or pcDNA3-Fyn-CA. Cells were fixed and incubated with the mouse Fyn monoclonal antibody. Immunofluorescence was performed using the Alexa Fluor 594 Anti-mouse IgG. (D) Percentage of C2C12 cells with pEGFP-LKB1 signal detected in the cytoplasm. Data are representative of n = 5 experiments. (E) Fully differentiated 3T3L1 adipocytes were cotransfected with pcDNA3-LKB1 and pcDNA3-Fyn-KD or pcDNA3-Fyn-CA. Immunofluorescence was performed using the rabbit LKB1 polyclonal antibody and mouse Fyn monoclonal antibody followed by Alexa Fluor 488 anti-rabbit IgG and Alexa Fluor 564 anti-mouse IgG. (F) Percentage of 3T3L1 cells with LKB1 signal detected in the cytoplasm. Data are representative of n = 5 experiments (see also Figures S2–S5). Cell Metabolism 2010 11, 113-124DOI: (10.1016/j.cmet.2009.12.010) Copyright © 2010 Elsevier Inc. Terms and Conditions

Figure 5 Fyn Phosphorylates LKB1 on Tyrosine Residues 261 and 365 (A and B) (A) Gastrocnemius muscle and (B) differentiated 3T3L1 adipocyte extracts were immunoprecipitated with IgG or the Fyn rabbit polyclonal antibody and immunoblotted with the monoclonal LKB1 antibody. 3T3L1 adipocytes were transfected with the pcDNA3 empty vector or pcDNA3-Fyn construct. (C) Cell extracts (lysates) were immunoblotted for Fyn and LKB1. (D) Cell extracts were immunoprecipitated with the LKB1 monoclonal antibody and immunoblotted with the phosphotyrosine antibody (PY100) or LKB1 antibody. (E) Purified His-tagged LKB1 was incubated with ATP in the absence and presence of purified Fyn protein. The samples were then immunoblotted with the LKB1 antibody and the phosphotyrosine antibody PY100. (F) pcDNA3-Flag-LKB1 mutant cDNAs and the pcDNA3-Fyn-CA constructs were coexpressed in 3T3L1 adipocytes, and levels of expression were determined in whole-cell extracts. (G) Levels of tyrosine phosphorylation of each LKB1 construct were determined in Flag immunoprecipitates subjected to immunoblotting with PY100. pcDNA3-Flag-LKB1-Y261/365F double mutant and pcDNA3-Fyn-CA constructs were coexpressed in 3T3L1 adipocytes. (H) The levels of LKB1 expression were determined by immunobloting cell extracts with the Flag and Fyn antibodies, respectively. (I) LKB1 tyrosine phosphorylation was determined by LKB1 immunoprecipitation followed by immunoblotting with the PY100 and Flag antibodies (see also Figure S6). Cell Metabolism 2010 11, 113-124DOI: (10.1016/j.cmet.2009.12.010) Copyright © 2010 Elsevier Inc. Terms and Conditions

Figure 6 LKB1 Tyrosine Phosphorylation Regulates Its Subcellular Distribution (A) 3T3L1 adipocytes were transfected with pcDNA-Flag-LKB1-WT or the pcDNA-Flag-LKB1-Y60F mutant cDNAs. (B) 3T3L1 adipocytes were transfected with the pcDNA-Flag-LKB1-Y261F and pcDNA-Flag-LKB1-Y365F cDNAs. (C) 3T3L1 adipocytes were transfected with pcDNA-Flag-LKB1-Y261/365F double mutant cDNA. Cells were fixed and subjected to immunofluorescence for the localization of LKB1 (Flag antibody) and DAPI labeling for nuclei identification. (D) 3T3L1 adipocytes were transfected with pcDNA-Fyn-CA and the pcDNA-Flag-LKB1-Y261/365F double mutant cDNAs. Cells were fixed and subjected to immunofluorescence for Fyn-CA expression (red), LKB1-Y261/365F double mutant localization (green), and nuclei (blue). (E) Percentage of 3T3L1 cells with LKB1 signal detected in the cytoplasm. Data are representative of n = 3 experiments. Cell Metabolism 2010 11, 113-124DOI: (10.1016/j.cmet.2009.12.010) Copyright © 2010 Elsevier Inc. Terms and Conditions

Figure 7 Subcellular Localization of LKB1 in Skeletal Muscle In Vivo Is Regulated by Tyrosine Phosphorylation (A) Tibialis anterior was transfected with pcDNA-Flag-LKB1-WT (Aa–Ac), the pcDNA-Flag-LKB1-Y261/365F double mutant (Ad–Af), or the pcDNA empty vector (Ag–Ai) cDNAs. Immunofluorescence was performed on 10 μm frozen sections for the localization of LKB1 (Flag antibody) and nuclei (DAPI). (B) Magnified images of muscles transfected with pcDNA-Flag- LKB1-WT (Ba–Bc) or the pcDNA-Flag-LKB1-Y261/365F double mutant (Bd–Bf) is shown to more easily visualize the change in LKB1 localization (see also Figure S7). Cell Metabolism 2010 11, 113-124DOI: (10.1016/j.cmet.2009.12.010) Copyright © 2010 Elsevier Inc. Terms and Conditions